Base mainnet down for two days due to sequencer bug

The Base mainnet experienced two separate outages on 25 and 26 June 2026, both caused by the same bug in the sequencer. The first outage began at 11:47 a.m. EDT on the 25th and lasted a full 116 minutes. The second outage followed a day later at 11:28 a.m. EDT and was resolved within 20 minutes. Base has now published a post-mortem in which the team outlines the cause, impact and measures taken.

The same bug, twice

Both incidents trace back to exactly the same cause: an error in the block-building process of the sequencer. The bug caused outdated journal state to persist after a failed transaction. As a result, the sequencer produced a block with an invalid state transition, which completely halted block production.

Base states that the integrity of the blockchain was never compromised. All funds on the network remained safe during both outages. Once the team identified the issues and intervened, block production resumed without further complications.

Improvements to prevent recurrence

Now that the bug itself has been fixed, the Base team is also looking at broader improvements in the development process. Specifically, there are plans to strengthen protocol fuzz testing and load testing. Additionally, the team wants to improve monitoring and make network recovery mechanisms more robust.

The fact that both outages were caused by exactly the same bug indicates that the fix after the first outage was insufficient to prevent recurrence. The team acknowledges this and aims with the planned improvements to prevent similar problems from occurring again in the future.